Advent Calendar for ML and DL

Towardsdatascience launched a December Advent Calendar "Machine and Deep Learning", which offers to understand what is under the hood of ML processes. Frameworks like scikit-learn have made us lazy. The call to model.fit has become so commonplace that in the era of Gen AI, it seems like training a model is just parameter tuning. ML engineers juggle models with complexity that grows exponentially, yet they are not always able to manually recalculate and explain the results of even the simplest algorithms: linear regression or classifiers. Models have turned into "black boxes", and this is a huge problem, as knowing what is behind each function is critically important for understanding the process. The point is that all material is analyzed in Excel. It sounds strange, but therein lies the genius. Unlike code, where operations are hidden behind functions, in Excel every formula, every number, every calculation is all visible. No "black boxes". Seven articles have already been released: Day 1: k-NN Regressor Day 2: k-NN Classifier Day 4: GNB, LDA, and QDA Day 5: GMM (Gaussian Mixture Model) Day 6: Decision Tree Regressor Day 7: Decision Tree Classifier The series will help answer questions that often remain behind the scenes: how to properly handle categorical features, when scaling is not the right solution, and how to measure feature importance by interpreting them directly with the model, bypassing model-agnostic packages like LIME and SHAP. The series will be useful for students to understand formulas, and for managers to comprehend which ML method is necessary for business. For developers, it is a chance to finally understand the theory. In general, this is a must-read for those who want to stop being library operators and truly understand how the ML engine works. You can monitor the release of new articles here, with a promise of publication by the end of December in the format "one day - one article."